On discovering co-location patterns in datasets: a case study of pollutants and child cancers

摘要: We intend to identify relationships between cancer cases and pollutant emissions by proposing a novel co-location mining algorithm. In this context, we specifically attempt understand whether there is relationship the location of child diagnosed with any chemical combinations emitted from various facilities in that particular location. Co-location pattern intends detect sets spatial features frequently located close proximity each other. Most previous works domain are based on transaction-free apriori-like algorithms which dependent user-defined thresholds, designed for boolean data points. Due absence clear notion transactions, it nontrivial use association rule techniques tackle problem. Our proposed approach focused grid transactionization? geographic space, mine datasets extended objects. It also capable incorporating uncertainty existence model real world scenarios more accurately. eliminate necessity using global threshold introducing statistical test validate significance candidate patterns rules. Experiments both synthetic reveal our algorithm can considerable amount statistically significant patterns. addition, explain modelling framework used pollutants (PRTR/NPRI) childhood cases.